As a speaker, I sometimes receive the results of survey forms completed by conference delegates. Mostly, they seem to like what they hear. They give high marks for relevance, presentation, originality. But some people think you are very, very bad. Not just irrelevant, but poor in presentation, weak in delivery and originality. Could these people really have heard the same speech?
Some people like you, others don't. The ones who like you give you high marks for everything, and the ones who don't, give you low marks for everything. It's rare for someone to say the material was right but the presentation didn't quite come off, or that I sounded convincing even though I didn't really have anything new to say.
The 'halo effect' was discovered a century ago in the assessment of military capability. Recruits were assessed as having all soldierly qualities or none. Few people were good in some respects, deficient in others. When hotel guests fill in comment forms, those who think the room was spacious also find the reception staff friendly and the restaurant food good. People who think their room is too small tend to find the employees surly and the breakfast disappointing.
Phil Rosenzweig describes how the halo effect does not simply colour accounts of business performance, it dominates them. When Cisco was thought to be performing well in the 1990s, the company was admired for its customer focus, its ability to integrate acquisitions, and for the rigour of its cost control. Suddenly, in mid-2000, all these judgments were revised. The firm had been cavalier in its treatment of customers, was a sprawling conglomeration of disparate businesses, and its systems had failed to give management the required information. The halo had become a cloven hoof. And, Rosenzweig emphasises, no-one was saying Cisco had changed.
Similarly, Percy Barnevik was for years Europe's most admired business leader. Then, as ABB stumbled, business journalists and academics told a different story. What had seemed to be charisma was a cult of personality. The most upright corporate citizens had engaged in questionable practices. Risks that once seemed so finely judged had been rash gambles.
It wasn't that Barnevik had lost the plot. It was just that the CEO of a company that exceeds expectations is described differently from one whose firm is underperforming. Even when the CEO is the same individual.
Rosenzweig offers a telling indictment of the superficiality of much management literature, and a critique of the many studies that claim to illuminate the source of high performance through the experience of successful companies - the literature that has burgeoned since Tom Peters' In Search of Excellence. Insights into corporate success are often only restatements that the company is doing well.
So successful firms have customer focus and tight cost control? That's really useful information for people who thought it was a bad idea to focus on customers and a good one to give up cost control, but most managers have reached that insight for themselves. Successful companies bask in the halo effect conferred by having got the basics of management right. But it doesn't tell us how they did it - or, crucially, how they did it more successfully than their competitors.
More problematically still, these characteristics of success are often correlated. A hotel that has friendly staff needn't have pristine bedrooms, but it probably does: a management good at inspiring the front desk will often be successful in motivating the chambermaids.
There is a large random element in performance. Cisco was probably never particularly good or bad. It had the good luck to live through a period - the late 1990s - when its customers' demand for its products was insatiable. Then the wheel of fortune turned. The business didn't change; the environment in which the business operated did.
The man who buys the winning ticket in a lottery will always be admired for his shrewdness. Just don't ask him for tips on how to win the lottery next time. Asked for the secret of his success, Paul Getty reportedly said: 'Strike oil', and he got to the heart of the matter, even though it wasn't the kind of answer his interlocutor was looking for.
Books like In Search of Excellence and Built to Last are great stories, with genuine insights from thoughtful observers of management practice. But the claimed rigour of their scientific analysis isn't worth a row of beans, and Rosenzweig explains why with wit and vigour. Put The Halo Effect alongside these earlier volumes on your business bookshelf. The Halo Effect ... and the eight other business delusions that deceive managers
Simon & Schuster £16.99
John Kay is an economist and author of The Hare and the Tortoise (Erasmus Press).